Slingshot RL framework jailbreaks Qwen2.5-32B at 67% success, transfers zero-shot to Gemini 2.5 Flash

j_foerst · x · 2026-09-23

Paper: David vs. Goliath — Verifiable Agent-to-Agent Jailbreaking

Researchers Samuel Nellessen and Tal Kachman (arXiv 2602.02395) formalize Tag-Along Attacks: a tool-less adversary rides on a safety-aligned Operator's legitimate tool privileges to induce prohibited tool use through conversation alone, turning tool-environment safety evaluation into an objective control problem.

Slingshot framework (cold-start RL):

The work establishes Tag-Along Attacks as a first-class verifiable threat model, showing effective agentic attacks can be autonomously discovered from off-the-shelf open-weight models.

Original post →

More from Safety

Safety channel →